The proposed research would study the effects of disasters on the mental health of older adults. It would test specific hypotheses about the mediating roles of disaster-induced life events and of changes in personal resources, such as social support, and about the duration and timing of disaster effects. These are important but difficult questions for disaster research because they require a prospective design. An unusual opportunity exists, however, because of the concurrence of a serious flood and a longitudinal study of resources, stress, and mental health of older persons. In June, 1981, widespread flooding caused $9 million in damage and left over 500 families homeless in eastern Kentucky, a relatively impoverished area that is at risk for future flooding. This flood occurred immediately after the first wave of data collection of a panel study in which older adults across Kentucky were interviewed up to five times at six-month intervals. 243 panelists lived within flooded counties and participated in both the interview wave that preceded the flood and the first interview wave that followed the flood. Other floods that occurred during the course of this study yield an additional 70 respondents in eastern Kentucky that have both pre- and post-flood measures. In sum, this data provides pre and post measures of resources, life events, and mental health for a high-risk sample that was not selected because of their flood exposure. We would define the independent variable (flood exposure) according to two criteria: impact and recurrence. For both constructs, we would obtain objective measures, drawn from administrative records, and self report measures, obtained by reinterviewing panelists who lived in flooded or adjacent counties. Objective measures would include county property damage and frequency of floods in that county. Parallel self report measures would be personal property damage and previous exposure to traumatic events. We also would obtain new data about current mental health, resources, and perceived risk which would allow the long range impact of flood exposure to be examined.